Automated 3D lymphoma lesion segmentation from PET/CT characteristics

Abstract

International audiencePositron Emission Tomography (PET) using 18 F-FDG is recognized as the modality of choice for lymphoma, due to its high sensitivity and specificity. Its wider use for the detection of lesions, quantifica-tion of their metabolic activity and evaluation of response to treatment demands the development of accurate and reproducible quantitative image interpretation tools. An accurate tumour delineation remains a challenge in PET, due to the limitations the modality suffers from, despite being essential for quantifying reliable changes in tumour tissues. Due to the spatial and spectral properties of PET images , most methods rely on intensity-based strategies. Recent methods also propose to integrate anatomical priors to improve the seg-mentation process. However, the current routinely-used approach remains a local relative thresholding and requires important user interaction , leading to a process that is not only user-dependent but very laborious in the case of lymphomas. In this paper, we propose to rely on hierarchical image models embedding multimodality PET/CT de-scriptors for a fully automated PET lesion detection / segmenta-tion, performed via a machine learning process. More precisely, we propose to perform random forest classification within the mixed spatial-spectral space of component-trees modeling PET/CT mages. This new approach, combining the strengths of machine learning and morphological hierarchy models leads to intelligent thresholding based on high-level PET/CT knowledge. We evaluate our approach on a database of multi-centric PET/CT images of patients treated for lymphoma, delineated by an expert. Our method provides good efficiency, with the detection of 92% of all lesions, and accurate seg-mentation results with mean sensitivity and specificity of 0.73 and 0.99 respectively, without any user interaction

Deep learning enables automatic detection and segmentation of brain metastases on multisequence MRI

“…During recent years, advances in machine learning (ML) have suggested the possibility of new paradigms in healthcare. One application of ML in radiology is the detection and segmentation of organs and pathology . In particular, there has been a significant effort in developing deep learning (DL) algorithms to learn from the comprehensive voxelwise labeled MRI data for segmenting primary brain tumors .…”

Shaping for PET image analysis

“…A unique criterion, combining intensity and compacity priors, may not be sufficiently specific and representative for all lymphoma lesions. Therefore, considering machine (Grossiord and et al, 2017) or deep learning procedures on morphological hierarchy models could help rely on relevant image descriptors that capture physiological phenomena more comprehensively.…”

“…Relying on the fact that high metabolic activity regions appear as hyperintense areas in PET images, methods were designed to take advantage of the mixed spatialspectral organization of PET information in componenttrees to develop classification strategies (Alvarez Padilla and et al, 2015;Grossiord and et al, 2017) or filtering / segmentation methods (Urien and et al, 2017;Alvarez Padilla and et al, 2018a,b). Indeed, PET visualizes metabolic activity characterized by the intensity of an injected radiotracer.…”

Hierarchical forest attributes for multimodal tumor segmentation on FDG-PET/contrast-enhanced CT

“…To improve robustness and ergonomy, some methods have been introduced [1]. They can be classified with respect to their methodology: thresholding [2,3], learning-based [4], boundary-based (level set [5], gradient-based [6]), statistical scopes (FLAB [7], fuzzyc-means [8], Random Walker [9]), mathematical morphology (watershed [10], component-tree [11]) and hybridization [12,13]. Most of these methods have been designed for processing PET images because of its high sensitivity and specificity to tumor biomarker metabolism.…”